HHS Public Access

Author manuscript
Child Psychiatry Hum Dev. Author manuscript; available in PMC 2022 December 01.
Author Manuscript

Published in final edited form as:

Child Psychiatry Hum Dev. 2022 December ; 53(6): 1266–1280. doi:10.1007/s10578-021-01205-7.

Longitudinal Associations between Parental Warmth, Harsh

Discipline, Child Emotion Regulation, and ODD Dimensions
Rachelle A. Yu,
Natalie Goulter,
Robert J. McMahon
Department of Psychology, Simon Fraser University, Burnaby, BC, Canada;
Author Manuscript

BC Children’s Hospital Research Institute, Vancouver, BC, Canada

Abstract
Research has yet to investigate the mediating role of emotion regulation in the relation between
parental warmth and harsh discipline with oppositional defiant disorder (ODD). Further, few
studies have investigated ODD as both a unitary construct and as three distinct dimensions (anger,
defiance, spitefulness). Thus, the present study aimed to investigate child emotion regulation
(grade 2) as a mediator of the relation between parental warmth and harsh discipline (kindergarten,
grade 1, grade 2) and ODD and its dimensions (grade 3). Participants included the high-risk
and normative samples from the Fast Track project (N=753, male=58%, Black=46%). Constructs
were assessed using observation and parent and teacher reports. Although results demonstrated an
Author Manuscript

absence of indirect effects, emotion regulation was negatively associated with overall ODD and
anger and defiance, but not spitefulness. Findings illustrate how increased attention toward the
study of ODD as distinct dimensions contributes to our ability to parse out heterogeneity among
children with the disorder.

Keywords
parental warmth; harsh discipline; emotion regulation; oppositional defiant disorder;
developmental psychopathology

Among the most commonly diagnosed childhood disorders, oppositional defiant disorder
(ODD) is characterized by a pattern of negativistic, hostile, and disobedient behaviors,
Author Manuscript

particularly toward authority figures [1]. Children with ODD frequently exhibit disruptive
behaviors including temper tantrums, argumentativeness, and unwillingness to compromise.
Epidemiological studies have consistently found that it is the most prevalent disorder in the

Mailing Address: Rachelle A. Yu, Simon Fraser University, 8888 University Drive, Burnaby, BC, V5A 1S6, [email protected], Telephone:
(778) 688-2894.
Conflicts of Interest: The authors declare that they have no conflicts of interest.
Ethical Approval. This study was approved by The Research Ethics Board of Simon Fraser University. All procedures performed
in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research
committee with the 1964 Helsinki Declaration and its later amendment or comparable ethical standards.
Informed Consent. Informed consent was obtained from all individual participants included in the study.
 Yu et al. Page 2

preschool years, with rates ranging from 6 to 13% among children aged 3–6 [2]. In addition,
Author Manuscript

ODD is associated with an increased risk for a host of negative outcomes in adolescence and
adulthood, including antisocial behavior, substance misuse, depressive disorders, and anxiety
disorders [3], as well as greater social impairment and greater family dysfunction [4]. Taken
together, these findings highlight the importance of studying risk factors of ODD symptoms
within this particular developmental period (i.e., early school years). Importantly, a large
body of research has investigated ODD within the context of broadband categories, such as
disruptive behavior disorders, externalizing problems, and/or conduct problems. To extend
this research, it is important for studies to examine whether established patterns in these
broader categories are distinct in diagnostic classifications, such as ODD and its constituent
components.

One focus of empirical research on ODD over the past decade has been on the dimensional
aspects of its diagnostic criteria. The strongest empirical support at present is for
Author Manuscript

three distinct yet interrelated dimensions: angry/irritable mood (anger), defiant/headstrong

behavior (defiance), and spitefulness/vindictiveness (spitefulness) [5], which has been
incorporated into the most recent edition of the Diagnostic and Statistical Manual of Mental
Disorders, 5th Edition (DSM-5) [1] in its description of ODD symptoms. Research has
demonstrated differential predictions of later psychopathology for both girls and boys,
with the anger dimension predicting internalizing psychopathology, and the defiance and
spitefulness dimensions predicting conduct disorder (CD) symptoms and delinquency [6].
The associations between ODD dimensions and differential outcomes have received support
from a number of longitudinal studies using community, high-risk, and clinic-referred
samples [7–10] and are identifiable with adequate reliability, convergent validity, and
discriminant validity as early as the preschool years [11]. In one study using class-based
analyses, the authors investigated the predictive validity of ODD irritable and defiant
Author Manuscript

dimensions in a large sample of children and youth aged 4–16 into adulthood. The
irritable dimension was associated with increased odds of adult mood disorders, whereas
the defiant dimension was associated with increased odds of violence as adults [12]. A
cross-sectional study examining differential correlates of a two-dimensional breakdown of
ODD in girls and boys (ages 5–12) found that irritability was associated with physical
victimization and withdrawal, whereas defiance was associated with proactive aggression
and hyperactive-impulsive symptoms [13]. However, it is important to note that this study
did not identify spitefulness as a third dimension in the authors’ conceptualization of ODD.
This body of research points to the importance of investigating factors that may mitigate
children’s risk of developing ODD, particularly examining risk in relation to its distinct
dimensions. Researchers have made calls for studies to focus on ODD through dimensional
conceptualizations, rather than as a unitary construct [14]. The existing research on ODD
Author Manuscript

can be expanded by exploring predictors of the three-dimensional structure of ODD, and

potential differential associations with each distinct dimension, over time.

Parenting Behaviors and ODD

Parenting behaviors have been identified as important factors influencing the development
of disruptive behavior. Negative parenting behaviors, such as physical or verbal punishment,
intrusiveness, and physical punishment have consistently been found to be related to the

Child Psychiatry Hum Dev. Author manuscript; available in PMC 2022 December 01.
 Yu et al. Page 3

development of child externalizing problems (e.g., [15]). Much of the initial research on this
Author Manuscript

topic was conducted by Patterson [16], who theorized that a high rate of coercive exchanges
between the parent and child can train children to employ aggressive or oppositional
strategies to avoid parental demands, thereby gaining control over a disrupted or chaotic
home environment [17]. After intensive practice, these coercive patterns of parent-child
interaction can be generalized from within the family to other contexts (e.g., peers and
teachers), thereby giving rise to behavior problems that stabilize across settings and
throughout development [18].

Meta-analytic research (k = 1,435) with girls and boys has reported that higher levels of
harsh, punitive parenting behaviors are associated with increased externalizing symptoms,
while higher levels of parental warmth are linked to decreased externalizing symptoms (e.g.,
[15]). These associations have been found consistently in both childhood and adolescence.
Cross-sectional research has identified specific links between low parental warmth and
Author Manuscript

child oppositional behaviors for girls and boys in kindergarten (e.g., [19], using the same
sample as the current study) as well as harsh parenting and child ODD symptoms in
preschoolers [20]. Early research suggested that the cultural context in which physical
punishment occurs is critical in determining the meaning of the discipline to the child
(e.g., spanking may represent good or bad parenting depending on the prevalence and
acceptance of the parenting behavior within the child’s broader cultural group) [21]. In
contrast, recent longitudinal and cross-cultural research has found that physical discipline
predicted elevations in child externalizing symptoms across racial and ethnic groups and
independent of cultural context [22].

In contrast to negative aspects of parenting (such as harsh discipline), which have been a
primary focus of research investigating predictors of child externalizing problems, positive
Author Manuscript

aspects of parenting (such as parental warmth) have received less attention as potential
protective mechanisms against the development of externalizing problems [15]. This is
surprising given that low parental warmth and harsh discipline are two parenting behaviors
that commonly co-occur. These parenting behaviors have been shown to predict child
externalizing problems longitudinally in both childhood and adolescence and across multiple
cultural groups and countries [23, 24, 25]. Longitudinal research has identified parental
alienation (defined as level of emotional isolation between the parent and child) as a
mediator in the relationship between family violence and child ODD symptoms [26].
Researchers have proposed that high levels of parental warmth are associated with better
adjustment, social competence, and academic readiness in preschool [27] while promoting
the attachment bond in the parent-child relationship [28]. Both parental warmth and harsh
discipline have been associated with child conduct problems in both girls and boys and are
Author Manuscript

intervention targets in parenting programs for child conduct problems (e.g., [29]).

Using the current study’s sample from the Fast Track project, three longitudinal studies have
investigated parental warmth and harsh discipline in relation to conduct problems, callous-
unemotional traits, and antisocial outcomes [30, 31, 32]. One cross-sectional study examined
parental warmth and punitive discipline and associations with oppositional behavior using
both the normative and high-risk sample from the Fast Track project [19]. More specifically,
results identified links between low parental warmth and oppositional behavior, as well

Child Psychiatry Hum Dev. Author manuscript; available in PMC 2022 December 01.
 Yu et al. Page 4

as physically aggressive parenting and aggressive child behavior. Using a non-clinical

Author Manuscript

preschool sample, cross-sectional research has also found significant associations between
harsh parenting and child ODD symptoms [20] In addition, a study conducted in a large
community sample (N = 2,582) of children in the 5th and 6th grades (i.e., slightly older
than children in the present study) found that high levels of maternal warmth were
linked to fewer disruptive behaviors while buffering against the detrimental effects of
harsh physical discipline [33]. Those findings replicated results from an earlier study
sampling both biological and adoptive mother-child dyads, which demonstrated that the
link between maternal harsh physical discipline and early child externalizing problems
was strongest when the mother-child relationship was low in warmth [34]. These results
held after controlling for the genetic similarity of the mother-child dyad (i.e., passive
gene-environment correlation). However, to our knowledge, few studies have explicitly
investigated differential associations between parental warmth and harsh discipline and
Author Manuscript

ODD within the same model [19], constituting a significant limitation of existing research
given the high covariance between these two parenting behaviors. In addition, the majority
of studies have examined ODD as a unitary diagnostic category; thus, further research is
needed to determine how parental warmth and harsh discipline may be associated with
distinct ODD dimensions.

Child Emotion Regulation and ODD

Given the core symptomology of ODD (i.e., recurrent outbursts, persistent resistance), it
is reasonable to conjecture that children with ODD struggle significantly with emotion
regulation. Although it has been less frequently studied, deficits in emotion regulation
have been implicated as an etiological pathway to the development of ODD in childhood
[35, 36]. Emotion regulation is defined here as children’s ability to monitor and modulate
Author Manuscript

their emotional experience through either independent or interpersonal means [37] for the
purposes of achieving a goal [38]. Emotion regulation can include self-soothing, reframing
upsetting events [39], and/or inhibiting or initiating emotionally driven behavior [40].
Emotion regulation is deemed adaptive when the child is successful in altering an emotional
trajectory that may have occurred in the absence of a particular emotion regulation strategy
[38]. Researchers have identified emotion regulation as a transdiagnostic marker that
underlies many forms of psychopathology [41, 42, 43]. Further research is needed to
investigate whether emotion regulation is a mechanism that also underpins each of the three
dimensions of ODD.

The demands of the school environment often require these competencies from children,
especially as young children enter into the early school years (e.g., kindergarten). Research
Author Manuscript

demonstrates that emotion regulation skills develop between 2–5 years of age [44]. This
suggests that the early school years may be a critical period in which to study these
processes as the classroom environment begins to place demands on children’s capacities
for emotion regulation. This is particularly salient for children with ODD, as their behavior
problems tend to emerge during the same approximate time period [1, 45], and whose
emotion dysregulation can be especially disruptive in the classroom setting [46]. Although
some research has posited that ODD may be better conceptualized as part of a broader
disorder of emotion dysregulation for school-aged children [35] and pinpointed difficulties

Child Psychiatry Hum Dev. Author manuscript; available in PMC 2022 December 01.
 Yu et al. Page 5

in regulating emotional arousal as a precursor to ODD [36], these studies commonly

Author Manuscript

examine ODD as a unidimensional construct. Research has yet to explore the role of
emotion regulation in relation to individual ODD dimensions, representing an important
limitation.

Parenting Behaviors and Child Emotion Regulation

Emotion regulation skills are largely acquired through the family context, and more
specifically, through the parent-child relationship. Research has emphasized the importance
of key mechanisms for children’s development of emotion regulation skills including
children’s observations of parents’ emotion regulation (e.g., modelling, coaching), emotion-
related parenting practices (e.g., reactions to emotions), and the emotional climate of
the family (e.g., security and affective tone of the parent-child relationship, emotional
expressiveness) [47, 48]. Parents who demonstrate negative and hostile emotions may be
Author Manuscript

modeling dysregulated behavior to their children [40], a pattern of interaction that has
been associated with decreased emotional security for the child [49]. A review of evidence
gathered over the last 75 years indicated that parenting behavior is a vital process that
can either support or hinder children’s self-regulation through intergenerational mechanisms
[50]. Research has shown that the development of adaptive child emotion regulation skills
may be negatively impacted by low levels of maternal warmth and responsiveness to distress
[51, 52]. Depending on how emotion regulation skills have been modeled by the family
environment and internalized by the child, emotion regulation strategies may be transferred
from the home to the school and impact relationships with teachers and peers [53].

A small body of research has identified associations among the parenting behaviors of
interest in this study (i.e., parental warmth and harsh discipline), child emotion regulation,
Author Manuscript

and externalizing outcomes. Hostile parenting and maternal lack of sensitivity have been
associated with poor child emotion regulation, which then predicted more social and
behavioral problems in kindergarten and grade 1 [54]. However, to our knowledge,
longitudinal research has yet to examine the role of emotion regulation in the relationship
between parental warmth and harsh discipline and ODD, specifically. Researchers have
examined direct and indirect risk pathways between parenting factors (e.g., emotion beliefs
and behaviors, emotional expressiveness, mental health) and disruptive behavior disorders
in children [55]; however, this research did not examine ODD as a distinct outcome. This
represents a critical gap given that numerous studies have explored links between parenting
and emotion regulation, and between emotion regulation and ODD, yet no indirect models
have been undertaken to further investigate these pathways.
Author Manuscript

Current Study
Although it is well established that parenting behaviors are associated with the development
of child ODD symptoms, the mechanisms underlying this link have received less attention.
Thus, the current study employed a longitudinal design to examine the role of child emotion
regulation (grade 2) in the association between parenting behaviors (kindergarten, grade 1,
and grade 2) and ODD symptoms (grade 3). Two particular forms of parenting behaviors
(i.e., parental warmth and harsh discipline) were examined. Specifically, it was hypothesized

Child Psychiatry Hum Dev. Author manuscript; available in PMC 2022 December 01.
 Yu et al. Page 6

that child emotion regulation would mediate the relationship between parental warmth and
Author Manuscript

ODD, as well as the relationship between parental harsh discipline and ODD. An additional
aim of the study was to explore potential differential outcomes with respect to each
dimension of ODD (i.e., angry/irritable mood, argumentative/defiant behavior, spitefulness/
vindictiveness) as a means of advancing the conceptualization of the disorder, given that a
majority of the research has tended to examine ODD as a unitary diagnostic category.

To reduce inflated estimates due to common-method variance, constructs were assessed with
different reporters at each time point, with observer ratings of parental warmth and harsh
discipline at school entry (kindergarten, grade 1, and grade 2); parent and teacher ratings of
emotion regulation during middle childhood (grade 2); and parent reports assessing ODD
outcomes during middle childhood (grade 3).

Method
Author Manuscript

Participants
Participants came from the Fast Track project, a longitudinal, multisite investigation of the
development and prevention of childhood conduct problems [56]. Target schools within four
sites (Durham, NC; Nashville, TN; Seattle, WA; and rural Pennsylvania) were identified
as high risk based on crime and poverty statistics of the neighborhoods that they served.
From 1991–1993, 9,594 kindergarteners were screened using a multiple-gating screening
procedure that combined teacher and parent ratings of disruptive behavior. Children were
first screened for classroom conduct problems by teachers, using the Teacher Observation
of Classroom Adaptation-Revised Authority Acceptance score [57], and a subset of these
children were then screened for behavior problems at home using a 22-item parent-report
measure based on the Child Behavior Checklist [58]. The teacher and parent screening
Author Manuscript

scores were standardized and summed to yield a total severity-of-risk screen score. Children
were selected for inclusion into the high-risk sample based on this screen score, moving
from the highest score downward until desired sample sizes were acquired across sites,
cohorts, and groups. Overall, 891 children (control = 446 and intervention = 445) were
selected for the high-risk sample and participated in the study. In addition, a stratified
normative sample of 387 children was selected and followed over time to represent the
population normative range of risk scores.

The current study utilized data from 446 high-risk control (65% male; 44% Black, 51%
White, 5% other race) and normative (51% male; 42% Black, 51% White, 7% other race)
participants. Because 79 of those recruited for the high-risk control group were also included
as part of the normative sample and 1 participant had a missing value on a weighting
Author Manuscript

variable, the final analyzed sample size included 753 participants. At the first year of
data collection, the mean age of participants was 6.38 (SD = 0.45) years. According to
Hollingshead’s [59] rating system, the mean socioeconomic status (SES) continuous code
was 27.35 (SD = 10.92), indicating primarily lower to middle SES. Probability weights
were constructed to adjust for the over-sampling of high-risk children [60]. Each summer,
two research interviewers visited each home; one interviewed the primary caregiver and
the other interviewed the child. The interviewer read all questionnaires and recorded the
caregiver’s response. Computer-assisted interviews were conducted with the children, who

Child Psychiatry Hum Dev. Author manuscript; available in PMC 2022 December 01.
 Yu et al. Page 7

wore headphones and listened to items being read to them on a laptop computer, answering
Author Manuscript

directly. Parents were compensated with $75 for completing interviews each summer, and
teachers were compensated each year with $10 per child that they rated on classroom
measures. Informed consent was obtained from all individual participants included in the
study.

Measures and Procedures

Covariates.—Covariates measured in kindergarten included initial risk screen scores
summed from standardized teacher and parent screening scores (M = 1.10, SD = 1.64,
range = −3 to 5), sex (male = 58%), SES (M = 25.65; SD = 12.90) [59], and race/urban
status (urban Black = 45.5%, urban White = 24.5%, rural White = 25.5%). The race/urban
status variable was created to account for the multisite sampling of the Fast Track project
that resulted in almost all Black participants living in urban areas. Covariates were selected
Author Manuscript

based on existing research indicating that these are important factors for the development of
ODD.

Parenting measures.—A breakdown of the data collected from informants based on

primary caregiver relationship reported at kindergarten are as follows: 95.4% biological
mothers and other female caregivers (e.g., foster mother, adoptive mother, stepmother, other
female relative); 4.3% biological fathers and other male caregivers (e.g., stepfather, other
male relative).

Parental Warmth.—Parent-child dyads completed the Parent-Child Interaction Task

(PCIT) [61] during home visits occurring in the summers following kindergarten, grade
1, and grade 2. The PCIT is a semi-structured observation procedure consisting of 5 min
of free play, 5 min of parent-directed play, 5 min of a Lego building task, and 3 min of
Author Manuscript

clean-up. After each task, coders made ratings of parent and child behavior and affect using
an adaptation of the Interaction Rating Scale [62]. This measure is based on 16 global
items assessing gratification, sensitivity, and involvement using a global 5-point system (1
‘no enjoyment’, ‘intrusive’, ‘little interest’; 5 ‘long periods of enjoyment’, ‘no intrusions’,
‘predominantly engaged’, respectively). Scores for parental warmth were determined by the
mean of six items from gratification (e.g., enjoyment in the interaction with the child),
sensitivity (e.g., sensitive responsiveness to the child’s cues), and involvement (e.g., time
spent interacting with the child) that were coded across the first three tasks (Cronbach’s α
for kindergarten [Normative = 0.88, Control = 0.88], grade 1 [Normative = 0.88. Control =
0.90], and grade 2 [Normative = 0.90, Control = 0.92]).

At each site, coders received training on the observation systems by a lead observer. Coders
Author Manuscript

met on a weekly basis to minimize coder drift, and lead observers were trained annually and
participated in regular conference calls to control for cross-site coder drift.

Parental Harsh Discipline.—The Life Changes interview measure [63] was administered
to parents at the end of kindergarten, grade 1, and grade 2. One portion of this interview
assessed parental discipline strategies by providing parents with six brief written vignettes
detailing various episodes of child misbehavior (e.g., hitting another child, noncompliance).

Child Psychiatry Hum Dev. Author manuscript; available in PMC 2022 December 01.
 Yu et al. Page 8

Parents were asked what they would do in each scenario, and their responses were
Author Manuscript

coded into one of eight mutually exclusive categories. More specifically, coders rated
whether parents mentioned using any of the following discipline strategies: contingent
ignoring, doing nothing, reasoning/inductive reasoning/proactive guidance, inappropriate
verbal correction, appropriate verbal correction, withdrawal of privileges, directives, or
physical punishment. The Physical Punishment scale was used in this study, with the overall
score generated by an average of parents’ responses to the six vignettes (Cronbach’s αs
for kindergarten [Normative = 0.60, Control = 0.55], grade 1 [Normative = 0.47, Control =
0.42], and grade 2 [Normative = 0.44, Control = 0.30]).

Emotion Regulation.—Emotion regulation was assessed with parent and teacher versions
of the Social Competence Scale [SCS; 64]. The SCS – Teacher Version is a 25-item
measure assessing a child’s prosocial behavior, emotion regulation, and academic abilities
(Cronbach’s α = 0.96) that has been used with community, high-risk, and clinical samples
Author Manuscript

[65; Cronbach’s α = .78) in addition to the current sample [66; parents, α = .81, teachers,
α = .78; 67, parents, α = .85, teachers, α = .97]. In contrast to the parent version (see
below), the overall teacher-report scale consists of items describing behaviors that may be
displayed by the child, specifically within the school setting. Given that teachers are likely to
have broader experiences working with children and thereby a more informed understanding
of developmentally appropriate socioemotional functioning than parents, the teacher-report
version of this measure was included to capture emotion regulation difficulties exhibited by
the child at school. Sample items include ‘expresses needs and feelings appropriately’ and
‘can calm down when excited or all wound up’. Administered in the Spring of grade 2, the
teacher rated how well each item described the child on a 5-point Likert scale: “not at all
(0),” “a little (1),” “moderately well (2),” “well (3),” and “very well (4).” Three independent
subscales comprise the SCS–Teacher Version: Prosocial/Communication Skills, Emotion
Author Manuscript

Regulation Skills, and Academic Skills. This study utilized the Emotion Regulation Skills
subscale, which consists of 10 items assessing the degree to which a child exhibits emotional
and behavioral control in a social school setting (Cronbach’s αs grade 2: Normative = 0.96,
Control = 0.94). The Emotion Regulation Skills subscale was highly correlated with the
Prosocial Behavior (r= 0.91) and Academic Skills (r= 0.85) subscales.

The SCS – Parent Version is a 12-item measure assessing children’s prosocial skills,
communication skills, and self-control. Items are similar to the teacher-report version of this
subscale; sample items include “can accept things not going his/her way” and “copes well
with failure.” The parent assessed how well each item described a child on a 5-point Likert
scale: “not at all (0),” “a little (1),” “moderately well (2),” “well (3),” and “very well (4).”
Two independent subscales comprise the SCS–Parent Version: Prosocial/Communication
Author Manuscript

Skills and Emotion Regulation Skills. The Emotion Regulation Skills subscale consists of
six items assessing the degree to which a child exhibits emotional and behavioral control in
a social setting (Cronbach’s αs at grade 2: Normative = 0.82, Control = 0.76). The Emotion
Regulation Skills subscale was highly correlated with the Prosocial/Communication Skills (r
= 0.72) subscale.

Child Psychiatry Hum Dev. Author manuscript; available in PMC 2022 December 01.
 Yu et al. Page 9

In order to reduce inflated estimates between parent reports on emotion regulation and ODD,
Author Manuscript

all items across the teacher and parent versions of these measures were mean scored to be
included as the mediating variable.

ODD Symptom Dimensions.—In grade 3, children were assessed for ODD symptoms
using the Diagnostic Interview Schedule for Children (C-DISC) – Parent Report [68]. The
C-DISC is a highly structured computerized interview that assesses DSM-IV psychiatric
symptoms and diagnoses in children and adolescents aged 6–17, and is completed by the
parents (Cronbach’s αs at grade 3 ODD total = 0.86; anger subscale = 0.75; defiance
subscale = 0.70; spitefulness subscale = 0.71). Parents were asked whether their children
demonstrated symptoms related to the disorder within the past 6 months, and additional
follow-up questions if parents endorsed their children as having exhibited a given symptom.
Diagnostic criteria are determined by the C-DISC scoring program (see Table 1). ODD
symptom scores were based on 8 DSM-IV diagnostic criteria derived from 11 symptom
Author Manuscript

items in the measure. The selection and inclusion of C-DISC items into each of the three
dimensions was informed by previous research on the three-dimensional structure of ODD
[69, 70].

Data Analyses
Descriptive Analyses.—Descriptive analyses were conducted using SPSS version 26.0
[71] and structural models were conducted using Mplus version 7.1 [72]. Full information
maximum likelihood (FIML) with robust standard errors was used to address missing data,
which generates estimates of the variance-covariance matrix for all available data, including
participants with missing data on some measures [73]. Missing data are as follows: parental
warmth in kindergarten = 0.4%, grade 1 = 6.4%, grade 2 = 12.6%; harsh discipline in
Author Manuscript

kindergarten = 0.1%, grade 1 = 5.7%, grade 3 = 9.2%; emotion regulation in grade 2 =

2.1%; ODD at grade 3 = 12.7%, anger at grade 3 = 12.7%, defiance at grade 3 = 12.7%,
spitefulness at grade 3 = 13.1%. Model fit criteria were comprised of chi-square (χ2) values,
Root Mean Square Error of Approximation (RMSEA), and Comparative Fit Index (CFI) for
all models. Models were deemed an adequate fit to the data provided that a nonsignificant
χ2, RMSEA less than 0.06, and CFI greater than 0.90 were obtained [74].

Preliminary Analyses.—First, confirmatory factor analyses were estimated in relation to

parental warmth and harsh discipline as separate models without covariates [75]. Parental
warmth was measured across kindergarten, grade 1, and grade 2 as indicated by scores on
the IRS. Similarly, harsh discipline was measured across the same time points as indicated
by scores on the Physical Punishment scale. Secondly, an unconditional conjoint model
including both harsh discipline and parental warmth was analyzed. This conjoint model
Author Manuscript

(with parental warmth and harsh discipline modeled as latent factors) was used in all further
analyses.

Structural and Indirect Effect Models Involving ODD Outcomes.—Next, ODD

outcomes were added into the model as predicted by the covariates (i.e., initial risk screen
score, SES, sex (male = 1), site, urban/rural status (urban Black = 1 vs. rural White = 0;
urban White = 1 vs. rural White = 0), latent factor parental warmth, latent factor harsh

Child Psychiatry Hum Dev. Author manuscript; available in PMC 2022 December 01.
 Yu et al. Page 10

discipline, and emotion regulation. Direct and indirect associations of emotion regulation
Author Manuscript

between parental warmth and harsh discipline, and ODD outcomes, were examined using
the product of the coefficients method with bootstrapping to obtain 95% confidence intervals
of the mediated effect [76]. Analyses were conducted using 10,000 bootstrapped estimates
of the indirect effect. The indirect effects were considered significant if the 95% confidence
interval of the indirect effect did not include 0. Exploratory analyses were then repeated
with all three ODD dimensions included in the model and with each ODD dimension as an
independent outcome variable.

Results
Bivariate Associations.
Descriptive statistics and correlations are presented in Table 2. The means and standard
deviations of main study variables for the total sample were generated for parental warmth
Author Manuscript

for kindergarten, grade 1, and grade 2; harsh discipline for kindergarten, grade 1, and grade
2; emotion regulation for grade 2 both with and without the temper item1; ODD total
symptoms for grade 3; the anger dimension for grade 3; the defiance dimension for grade
3; and the spitefulness dimension for grade 3. Skewness and kurtosis for the measures were
examined, and all of the scores fell within the generally accepted guideline of ±2 standard
deviations.

Parental warmth and harsh discipline were negatively associated across kindergarten, grade
1, and grade 2. Neither parental warmth nor harsh discipline were associated with ODD
outcomes, with the exception of harsh discipline with ODD total in kindergarten and harsh
discipline with anger in kindergarten and grade 1. Parental warmth and harsh discipline
were positively and negatively associated with emotion regulation, respectively. Emotion
Author Manuscript

regulation was negatively associated with all ODD outcomes. All three dimensions of ODD
demonstrated significant positive associations with each other.

Preliminary Analyses.
First, parental warmth and harsh discipline were estimated in two separate unconditional
models without covariates. These models were ‘just identified,’ suggesting that the number
of observed parameters was equal to the number of estimated parameters, and thus, model
fit could not be assessed. Second, an unconditional model using parental warmth and harsh
discipline was estimated conjointly. This model provided adequate fit to the data (χ2(8)
= 26.04, p = .001, CFI = .98, RMSEA = .06, 90% CI[.03, .08]). In addition, parental
warmth and harsh discipline were negatively associated with each other (β = −.02, B(SE) =
−4.78(.00), p < .001).
Author Manuscript

1All analyses were conducted both with and without the child temper item (“often loses temper”) on the Emotion Regulation
Subscale, which posed the potential for an artificially inflated association with ODD symptoms. No significant differences were found
(M = 1.96, SD = 0.77 with the item; M = 1.96, SD = 0.76 without the item), and therefore all analyses reported in this study include
this item.

Child Psychiatry Hum Dev. Author manuscript; available in PMC 2022 December 01.
 Yu et al. Page 11

Structural and Indirect Effect Models Involving ODD Outcomes.

Author Manuscript

The model with parental warmth and harsh discipline was then expanded to include
covariates, emotion regulation, and ODD outcomes. ODD outcomes were analyzed
separately as a total score and with each dimension as an individual outcome.

ODD Total Score.

In the first model, ODD outcomes were examined as a total score. This model provided
adequate fit to the data (χ2(36) = 144.12, p < .001, CFI = .92, RMSEA = .07, 90% CI[.05,
.08]). Neither parental warmth nor harsh discipline were found to be significant predictors
of ODD. No significant associations were found between parental warmth and emotion
regulation, or between harsh discipline and emotion regulation (see Table 3). Emotion
regulation was significantly negatively associated with ODD total score.
Author Manuscript

In terms of associations with the covariates, parental warmth was significantly associated
with lower levels of initial risk, higher among urban White children compared with rural
White children, and positively associated with SES. Harsh discipline was significantly
associated with higher levels of initial risk, lower among urban White children compared
with rural White children, and negatively associated with SES. Emotion regulation was
significantly associated with lower levels of initial risk, associated with being female,
and lower among urban Black children compared with rural White children. ODD was
significantly associated with higher levels of initial risk, and lower among urban Black
children compared with rural White children.

ODD Dimensions.
Exploratory analyses were conducted to determine whether the model generated differential
Author Manuscript

outcomes for each independent ODD dimension (i.e., all three dimensions were included in
the same model).

When examining each dimension as individual outcomes, the model again provided
adequate fit to the data (χ2(44) = 167.28, p < .001, CFI = .94, RMSEA = .06, 90% CI[.05,
.07]). Neither parental warmth nor harsh discipline were significantly associated with anger,
defiance, or spitefulness. Neither parental warmth nor harsh discipline were associated with
emotion regulation (see Table 3). A significant negative association was found between
emotion regulation and anger and with defiance. All three dimensions were significantly
associated with each other.

With regards to covariates, anger was significantly associated with higher levels of initial
risk and lower among urban Black children compared to rural White children. Defiance and
Author Manuscript

spitefulness were both significantly associated with higher levels of initial risk.

Indirect effects of emotion regulation.

The primary hypothesis investigating whether emotion regulation accounted for the
relationship between parental warmth and harsh discipline and ODD outcomes was
examined. As there were no significant associations between parenting dimensions and
emotion regulation, no significant indirect effects were found (see Table 4).

Child Psychiatry Hum Dev. Author manuscript; available in PMC 2022 December 01.
 Yu et al. Page 12

Discussion
Author Manuscript

Although extensive research has identified associations between parenting behaviors and
externalizing problems in children (e.g., [15]), there remains a gap in the research base
that directly addresses the roles of parental warmth, harsh discipline, and child emotion
regulation in the development of ODD. Thus, the purpose of the current study was to
examine whether parental warmth and harsh discipline in kindergarten through grade 2
predicted ODD outcomes in grade 3, and whether child emotion regulation in grade
2 mediated this relationship, among normative and high-risk control children from the
Fast Track project. In addition, exploratory analyses were conducted to investigate ODD
outcomes as three independent dimensions, as defined by the DSM-5 (i.e., angry/irritable
mood, argumentative/defiant behavior, spitefulness/vindictiveness).

Contrary to initial hypotheses, no indirect effects were found between parental warmth
Author Manuscript

and ODD outcomes through child emotion regulation, and no indirect effects were found
between parental harsh discipline and ODD outcomes through child emotion regulation.
These findings were attributed to the absence of associations between parental warmth and
harsh discipline with emotion regulation. However, emotion regulation was found to be
significantly negatively associated with ODD at both levels of statistical analyses (i.e., ODD
total and individual ODD dimensions).

Parenting Behaviors and ODD

Some research has found that low levels of parental warmth have been linked to the
development of oppositional and externalizing behaviors [19, 24], including research that
examined the same sample as the current study [19], whereas other research has found no
association [20]. In the present study, we did not find a significant association between
Author Manuscript

parental warmth and ODD (e.g., [27, 32]). Although two studies using the same sample
have also found no significant longitudinal relationships between parental warmth and CD
symptoms [30, 31], another study demonstrated cross-sectional links between low parental
warmth and child oppositional behavior [19]. Prior research also has more robustly shown
links between harsh discipline and child conduct problems during the early school-age
years (e.g., [21]). Contrary to expectations, results of the present study did not support an
association between parental harsh discipline and ODD. However, at the bivariate level,
harsh discipline showed significant positive correlations with ODD total (in kindergarten)
and anger (in kindergarten and grade 1). (In contrast, parental warmth was not significantly
correlated with ODD total or any dimension at any time point.) It is possible that coercive
exchanges involving harsh discipline may be more frequent when children are learning
to adjust to kindergarten, which is a time of considerable transition. Despite this, the
Author Manuscript

structural model did not show a significant association between these variables. This finding
is inconsistent with other studies using the same sample, which have reported significant
longitudinal associations between harsh discipline and CD symptoms (as opposed to ODD
in the current study) [30, 31].

This finding may be due to a suppressor effect, defined as “a variable that increases the
predictive validity of another variable (or set of variables) by its inclusion in a regression
equation” [77, p. 246], which frequently occurs in social scientific studies. In this case, a

Child Psychiatry Hum Dev. Author manuscript; available in PMC 2022 December 01.
 Yu et al. Page 13

significant association between two variables may be reduced after controlling for all other
Author Manuscript

variables in the structural model, such that the residual variance becomes nonsignificant
[77]. Despite nonsignificant findings, the current study is the first to include both
parental warmth and harsh discipline within the same model to examine their longitudinal
associations with ODD. This represents an advantage over past research that has tended to
investigate parenting behaviors in relation to other forms of psychopathology, such as other
conduct problems, CD, and depressed mood [23, 24, 55].

Child Emotion Regulation and ODD

The current findings support prior research demonstrating associations between child
emotion regulation and the development of ODD symptoms [35, 36]. Deficits in effective
emotion regulation may be a critical factor contributing to the core symptomology of
ODD. Findings from the current study demonstrated a significant negative relationship
Author Manuscript

between emotion regulation and ODD at both levels of analyses. Specifically, at the bivariate
level, emotion regulation was associated with ODD total score and all three individual
dimensions. In the structural models, emotion regulation was associated with ODD total
score and anger and defiance dimensions. Research has suggested that emotion regulation
is a transdiagnostic construct underpinning many forms of psychopathology and diagnostic
categories [41, 42, 43]. The early school years represent an important period in which
both the development of emotion regulation skills and ODD symptoms appear to emerge.
Our findings suggest that in addition to representing a transdiagnostic indicator of broad-
spectrum psychopathology and diagnoses, emotion regulation may also be important in the
development of multiple dimensions of ODD. The present study adds to a limited body of
research investigating emotion regulation as a mechanism contributing to the development
of ODD in children. In addition, although no indirect effects were found, the current study’s
Author Manuscript

investigation of ODD as both a unitary construct and three distinct dimensions extends prior
research by illustrating significant associations between child emotion regulation and ODD,
as well as the anger and defiance dimensions of ODD.

Parenting Behaviors and Child Emotion Regulation

No significant associations were found for either parental warmth or harsh discipline with
child emotion regulation. These findings are inconsistent with a large body of research
indicating that the development of effective emotion regulation skills may be (a) promoted
by high levels of maternal warmth [51] and (b) compromised by insensitive or harsh
parenting [54]. Although using both parent and teacher reports of child emotion regulation
presents an advantage over child reports with respect to reliability and incremental
validity concerns [78], biases and other confounding variables (i.e., parent and/or teacher
Author Manuscript

psychopathology, which was not controlled for in this study) may affect parent and teacher
ratings of child behavior [79].

The items comprising our measure of emotion regulation are believed to capture the
construct in a manner that aligns with our conceptualization. However, alternative measures
may be more comprehensive in examining emotion regulation. For example, the Children’s
Emotion Management Scales for Sadness [CSMS; 80], for Anger [CAMS; 80] and for
Worry [CWMS; 81] assess self-reported behaviors in direct response to specific emotions.

Child Psychiatry Hum Dev. Author manuscript; available in PMC 2022 December 01.
 Yu et al. Page 14

Given the age at which this variable was assessed in our model, children may not have
Author Manuscript

developed adequate insight to provide self-reports of emotion regulation. In addition, it was

important to our study design to incorporate multi-informant parent and teacher reports to
reflect child emotion regulation across settings. Ultimately, our measure is believed to assess
common aspects of emotion regulation that are reflected in other widely used parent-report
measures such as the Emotion Regulation Checklist (ERC; [82]; e.g., lack of flexibility,
mood lability, empathy, emotional self-awareness).

ODD Outcomes by Dimension (Anger, Defiance, Spitefulness)

With the conceptual and clinical utility of a dimensional breakdown of ODD supported by
previous research [5, 6, 9], ODD outcomes were examined as three distinct dimensions.
Significant relationships at the bivariate level were demonstrated between anger and
defiance, but not spitefulness, thereby replicating previous research describing ODD as
Author Manuscript

three separate yet correlated dimensions [5]. When examined as independent dimensions of
ODD, emotion regulation was found to significantly predict anger and defiance. This finding
adds to prior research by demonstrating that associations between the anger dimension
of ODD and increased risk for emotional disorders (e.g., depression and anxiety) in
later development [6–9] may be identified in earlier development with emotion regulation
contributing to the development of ODD. Angry or irritable mood may reflect maladaptive
emotion regulation strategies wherein emotionally driven behaviors are not effectively
inhibited or controlled [40]. In a large sample of children aged 7–10 meeting DSM-5
criteria for ODD, the anger dimension was positively associated with difficulties in peer
relationships, and higher levels of argumentativeness were associated with lower levels of
prosocial functioning [10].

There was no significant association between emotion regulation and the spitefulness
Author Manuscript

dimension of ODD. In the present sample, the spitefulness items and criterion were endorsed
less frequently (items = 85–110; criterion = 49) than endorsements for items and criterion
indicating anger (items = 102–235; criterion = 48–79) and defiance (items = 132–258;
criterion = 56–84). While anger and defiance were measured using three and four criteria,
respectively, spitefulness was assessed using only a single criterion. It is also possible that
the items assessing spitefulness (e.g., “Often tries to get others in trouble”) represented
behaviors higher in severity relative to items assessing anger (e.g., “Often grouchy or easily
annoyed”) and defiance (e.g., “Often bothers others on purpose”), and thus, may be less
developmentally prevalent for children in the early school years. In addition, given that
spitefulness has been shown to be a robust predictor of callousness [6], emotion regulation
skills (whether adaptive or maladaptive) may be more difficult to detect in children who
exhibit shallow affect or deficits in empathy. This finding illustrates how the continued study
Author Manuscript

of ODD as distinct dimensions contributes to our ability to parse out heterogeneity among
children with the disorder.

Strengths, Limitations, and Future Directions

The present longitudinal study investigated the impact of two specific types of parenting
behaviors (e.g., parental warmth and harsh discipline) across three early school-age years
on ODD outcomes, with child emotion regulation hypothesized as a mediator of this

Child Psychiatry Hum Dev. Author manuscript; available in PMC 2022 December 01.
 Yu et al. Page 15

relationship. Within the context of broader research examining similar variables, the present
Author Manuscript

study design demonstrated several strengths. This study is relatively novel in its analysis of
ODD as both a unitary construct and by dimensions. It also employed rigorous methodology
to examine multiple covariates and observational parenting measures, utilized data from both
parent and teacher reporters across home and school settings, and used a large, multisite
sample. In addition, the study’s longitudinal design contributed to a growing body of
research investigating associations of predictor variables related to ODD dimensions across
time ([6–10, 12–13].

However, several limitations must also be considered. Whereas parental warmth was
assessed through direct observation of the parent and child, the measures of parental
discipline, emotion regulation (in part), and ODD outcomes were based on parent report,
which may have inflated associations between these constructs. However, the measure
of emotion regulation also included teacher reports to account for different settings in
Author Manuscript

which inappropriate emotion is displayed [83]. It would be important for future research
to examine whether current findings replicate when the same method was used for both
parenting behaviors. In addition, future researchers may aim to use a measure of harsh
discipline demonstrating stronger psychometric properties than the subscale used in the
present study. Although supplemental analyses were conducted to account for the potential
for artificially inflated associations due to overlap on one item in the measures of emotion
regulation and ODD outcomes (i.e., child temper item; see1), there may be further items
tapping into similar constructs across these measures. Although it has been used in other
studies, the present study’s emotion regulation measure has not been widely used in broader
research investigating emotion regulation in childhood. It is possible that items comprising
the emotion regulation measure may be tapping into other constructs in addition to emotion
regulation, such as behavioral or inhibitory control. However, the measure was developed as
Author Manuscript

an assessment of emotion regulation and demonstrated acceptable psychometric properties.

The time points selected for this study were limited to when relevant measures were
administered in the Fast Track project. That is, the measures of parental warmth and
harsh discipline were only administered during kindergarten, grade 1, and grade 2, and
the teacher- and parent-report measures of emotion regulation overlapped only in grade 2.
ODD outcomes in grade 3 were selected for the purposes of longitudinal proximity to the
parenting behavior and emotion regulation variables in the dataset (as well, the measure of
ODD outcomes was first administered to the sample in grade 3).

Current research on ODD is limited in several ways. Researchers should continue to

investigate both the full dimension and the three-dimensional conceptualizations of ODD
in children and adolescents. In addition, future research should continue to examine
Author Manuscript

alternative factor structures of ODD, as this may provide a more nuanced understanding
of heterogeneity among children displaying ODD symptoms. Researchers may also consider
including environmental risk factors that may influence the development of ODD while
controlling for genetic influence (e.g., by assessing ODD symptoms in parents) [84].
Future research should also examine associations between parental warmth, harsh discipline,
emotion regulation, and ODD at other developmental periods, such as how parental
warmth and harsh discipline during childhood may be associated with emotion regulation
and ODD outcomes in adolescence. Finally, future researchers should continue to focus

Child Psychiatry Hum Dev. Author manuscript; available in PMC 2022 December 01.
 Yu et al. Page 16

on the influence of other parenting behaviors and characteristics within the home or
Author Manuscript

family environment on the development of ODD (e.g., supportive parenting, siblings

exhibiting aggressive behavior). Such research may help to further illuminate potential
mechanisms underlying associations between parenting behaviors and ODD, as well as
emotion regulation and ODD.

Summary
Childhood ODD is associated with a host of later psychopathology, and thus, research
investigating predictors of ODD is necessary to inform the early identification of risk
factors. Although certain parenting behaviors have been identified as contributing risk
for the development of emotion regulation and ODD, little research has examined both
parental warmth and harsh discipline and the mediating role of emotion regulation in the
relation between these parenting behaviors with ODD. Further, emotion regulation has been
Author Manuscript

established as a transdiagnostic marker underlying many forms of psychopathology, but it

is currently unclear whether emotion regulation is a mechanism that also underpins each
ODD dimension (anger, defiance, spitefulness). Using the normative and high-risk control
samples of the Fast Track project, the current longitudinal study tested whether emotion
regulation (grade 2) mediated the association between parental warmth and harsh discipline
(kindergarten, grades 1, 2) and ODD total and subscale scores (grade 3). Structural equation
modeling showed no indirect effects due to an absence of significant associations between
parenting behaviors and emotion regulation. However, emotion regulation significantly
predicted ODD total and anger and defiance dimensions (but not spitefulness). Present
findings provide novel insight into child emotion regulation and associations with the
angry/irritable mood and defiant/headstrong dimensions of ODD. In addition, our findings
provide further support for the usefulness of a multidimensional conceptualization of ODD
Author Manuscript

symptoms and suggest that children’s emotion regulation skills have important implications
for the development of angry mood and defiant behavior.

Acknowledgements:
This work used data from the Fast Track project (for additional information concerning Fast Track, see http://
www.fasttrackproject.org). We are grateful to the members of the Conduct Problems Prevention Research Group
(in alphabetical order, Karen L. Bierman, Pennsylvania State University; John D. Coie, Duke University; D. Max
Crowley, Pennsylvania State University; Kenneth A. Dodge, Duke University; Mark T. Greenberg, Pennsylvania
State University; John E. Lochman, University of Alabama; Robert J. McMahon, Simon Fraser University and B.C.
Children’s Hospital Research Institute; and Ellen E. Pinderhughes, Tufts University) for providing the data and for
additional involvement.

We are grateful for the collaboration of the Durham Public Schools, the Metropolitan Nashville Public Schools,
the Bellefonte Area Schools, the Tyrone Area Schools, the Mifflin County Schools, the Highline Public Schools,
and the Seattle Public Schools. We appreciate the hard work and dedication of the many staff members who
Author Manuscript

implemented the project, collected the evaluation data, and assisted with data management and analyses.

Funding:
The Fast Track project has been supported by National Institute of Mental Health (NIMH) Grants R18MH48043,
R18MH50951, R18MH50952, R18MH50953, R01MH062988, K05MH00797, and K05MH01027; National
Institute on Drug Abuse (NIDA) Grants R01DA016903, K05DA15226, RC1DA028248, and P30DA023026;
National Institute of Child Health and Human Development Grant R01HD093651; and Department of Education
Grant S184U30002. The Center for Substance Abuse Prevention also provided support through a memorandum of
agreement with the NIMH. Additional support for this study was provided by a B.C. Children’s Hospital Research
Institute Investigator Grant Award and a Canada Foundation for Innovation Award to Robert J. McMahon.

Child Psychiatry Hum Dev. Author manuscript; available in PMC 2022 December 01.
 Yu et al. Page 17

References
Author Manuscript

1. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders: DSM-5.
5th ed. Washington, DC: American Psychiatric Association; 2013
2. Lavigne JV, Gouze KR, Hopkins J, Bryant FB, LeBailly SA (2011) A multi-domain model of
risk factors for ODD symptoms in a community sample of 4-year-olds. J Abnorm Child Psychol
40:741–757
3. Nock MK, Kazdin AE, Hiripi E, Kessler RC (2007) Lifetime prevalence, correlates, and persistence
of oppositional defiant disorder: Results from the National Comorbidity Survey Replication. J Child
Psychol Psychiatry 48:703–713 [PubMed: 17593151]
4. Greene RW, Biederman J, Zerwas S, Monuteaux MC, Goring JC, Faraone SV (2002) Psychiatric
comorbidity, family dysfunction, and social impairment in referred youth with oppositional defiant
disorder. Am J Psychiatry 159:1214–1224 [PubMed: 12091202]
5. Stringaris A, Goodman R (2009a) Three dimensions of oppositionality in youth. J Child Psychol
Psychiatry 50:216–223 [PubMed: 19166573]
6. Stringaris A, Goodman R (2009b) Longitudinal outcome of youth oppositionality: Irritable,
Author Manuscript

headstrong, and hurtful behaviors have distinctive predictions. J Am Acad Child Adoles Psychiatry
48:404–412
7. Burke JD, Hipwell AE, Loeber R (2010) Dimensions of oppositional defiant disorder as predictors
of depression and conduct disorder in preadolescent girls. J Am Acad Child Psy 49:484–492
8. Kolko DJ, Pardini DA (2010) ODD dimensions, ADHD, and callous–unemotional traits as
predictors of treatment response in children with disruptive behavior disorders. J Abnorm Child
Psychol 119:713–725
9. Rowe R, Costello JE, Angold A, Copeland WE, Maughan B (2010) Developmental pathways in
oppositional defiant disorder and conduct disorder. J Abnorm Child Psychol 119:726–738
10. Wesselhoeft R, Stringaris A, Sibbersen C, Kristensen RV, Bojesen AB, Talati A (2019) Dimensions
and subtypes of oppositionality and their relation to comorbidity and psychosocial characteristics.
Eur Child Adolesc Psychiatry 28:351–365 [PubMed: 30003396]
11. Ezpeleta L, Granero R, de la Osa N, Penelo E, Domènech JM (2012) Dimensions of oppositional
defiant disorder in 3-year-old preschoolers. J Child Psychol 53:1128–1138
Author Manuscript

12. Althoff RR, Kuny-Slock AV, Verhulst FC, Hudziak JJ, van der Ende J (2014) Classes of
oppositional-defiant behavior: Concurrent and predictive validity. J Child Psychol Psychiatry
55:1162–1171 [PubMed: 24673629]
13. Evans SC, Pederson CA, Fite PJ, Blossom JB, Cooley JL (2016) Teacher-reported irritable and
defiant dimensions of oppositional defiant disorder: Social, behavioral, and academic correlates.
School Ment Health 8:292–304
14. Ghosh A, Ray A, Basu A (2017) Oppositional defiant disorder: current insight. Psychology
research and behavior management.
15. Pinquart M (2017) Associations of parenting dimensions and styles with externalizing problems
of children and adolescents: An updated meta-analysis. Dev Psychol 53:873–932. [PubMed:
28459276]
16. Patterson GR Coercive family process. Finland: Castalia; 1982
17. Dishion TJ, Patterson GR. The development and ecology of antisocial behavior in children and
adolescents. In: Cichetti D, Cohen DJ, editors. Developmental psychopathology: volume three:
risk, disorder, and adaptation. 2nd ed. Hoboken (NJ): John Wiley & Sons Inc.; 2015. p. 503–541
Author Manuscript

18. Granic I, Patterson GR (2006) Toward a comprehensive model of antisocial development: A

dynamic systems approach. Psychol Rev 113:101–131 [PubMed: 16478303]
19. Stormshak EA, Bierman KL, McMahon RJ, Lengua LJ, Conduct Problems Prevention Research
Group (2000) Parenting practices and child disruptive behavior problems in early elementary
school. J Clin Child Adolesc Psychol 29:17–29
20. Cruz-Alaniz Y, Bonillo Martin A, Jané Ballabriga MC (2018) Parents’ executive functions,
parenting styles, and oppositional defiant disorder symptoms: a relational model. Universitas
Psychologica. 17:39–48.

Child Psychiatry Hum Dev. Author manuscript; available in PMC 2022 December 01.
 Yu et al. Page 18

21. Deater-Deckard K, Dodge KA (1997) Externalizing behavior problems and discipline revisited:
Nonlinear effects and variation by culture, context, and gender. Psychol. Inq 8:161–175
Author Manuscript

22. Gershoff ET, Sattler KM, Ansari A (2018) Strengthening causal estimates for links between
spanking and children’s externalizing behavior problems. Psychol Sci 29:110–120 [PubMed:
29106806]
23. Gershoff ET, Grogan-Kaylor A (2016) Spanking and child outcomes: Old controversies and new
meta-analyses. J Fam Psychol. 30:453–469 [PubMed: 27055181]
24. Rothenberg WA, Lansford JE, Alampay LP, Al-Hassan SM, Bacchini D, Bornstein MH, Chang L,
Deater-Deckard K, Di Giunta L, Dodge KA, Malone PS (2020) Examining effects of mother and
father warmth and control on child externalizing and internalizing problems from age 8 to 13 in
nine countries. Dev Psychopathol. 32:1113–1137 [PubMed: 31865926]
25. Janssens A, Goossens L, Van Den Noortgate W, Colpin H, Verschueren K, Van Leeuwen K (2015)
Parents’ and adolescents’ perspectives on parenting: Evaluating conceptual structure, measurement
invariance, and criterion validity. Assessment. 22:473–489 [PubMed: 25225229]
26. Chen H, Lin X, Heath MA, Ding W (2020) Family violence and oppositional defiant disorder
symptoms in Chinese children: The role of parental alienation and child emotion regulation. J Fam
Author Manuscript

Soc Work. 25:964–972

27. Ruberry EJ, Klein MR, Kiff CJ, Thompson SF, Lengua LJ (2017) Parenting as a moderator of
the effects of cumulative risk on children’s social–emotional adjustment and academic readiness.
Infant Child Dev. 27:1–18
28. George MR, Cummings EM, Davies PT (2010) Positive aspects of fathering and mothering, and
children’s attachment in kindergarten. Early Child Dev Care 180:107–119 [PubMed: 20657740]
29. McMahon RJ, Forehand RL. Helping the noncompliant child: Family-based treatment for
oppositional behavior. 2nd ed. New York: Guilford Press; 2003
30. Pasalich DS, Witkiewitz K, McMahon RJ, Pinderhughes EE, Conduct Problems Prevention
Research Group (2016) Indirect effects of the Fast Track intervention on conduct disorder
symptoms and callous-unemotional traits: Distinct pathways involving discipline and warmth. J
Abnorm Child Psychol 44:587–597 [PubMed: 26242993]
31. Goulter N, McMahon RJ, Pasalich DS, Dodge KA (2019) Indirect effects of early parenting
on adult antisocial outcomes via adolescent conduct disorder symptoms and callous-unemotional
traits. J Clin Child Adolesc Psychol:1–13
Author Manuscript

32. Zheng Y, Pasalich DS, Oberth C, McMahon RJ, Pinderhughes EE (2017) Capturing parenting as
a multidimensional and dynamic construct with a person-oriented approach. Prev Sci 18:281–291
[PubMed: 27179538]
33. McKee L, Roland E, Coffelt N, Olson AL, Forehand R, Massari C, Jones D et al. (2007) Harsh
discipline and child problem behaviors: The roles of positive parenting and gender. J Fam Violence
22:187–196
34. Deater-Deckard K, Ivy L, Petrill SA (2006) Maternal warmth moderates the link between physical
punishment and child externalizing problems: A parent-offspring behavior genetic analysis. Parent:
Sci Pract 6:59–78
35. Cavanagh M, Quinn D, Duncan D, Graham T, Balbuena L (2017) Oppositional defiant disorder is
better conceptualized as a disorder of emotional regulation. J Atten Disord 21:381–389 [PubMed:
24481934]
36. Greene RW, Doyle AE (1999) Toward a transactional conceptualization of oppositional defiant
disorder: Implications for assessment and treatment. Clin Child Fam Psychol Rev 2:129–148
Author Manuscript

[PubMed: 11227071]
37. Thompson R (1994) Emotion regulation: A theme in search of definition. Monogr Soc Res Child
Dev. 59:25–52 [PubMed: 7984164]
38. Gross JJ (2015) Emotion regulation: Current status and future prospects. Psychol. Inq 26:1–26
39. Schwartz D, Proctor LJ (2000) Community violence exposure and children’s social adjustment in
the school peer group: The mediating roles of emotion regulation and social cognition. J Consult
Clin Psychol. 68:670–683 [PubMed: 10965642]

Child Psychiatry Hum Dev. Author manuscript; available in PMC 2022 December 01.
 Yu et al. Page 19

40. Eisenberg N, Gershoff ET, Fabes RA, Shepard SA, Cumberland AJ, Losoya SH et al. (2001)
Mother’s emotional expressivity and children’s behavior problems and social competence:
Author Manuscript

Mediation through children’s regulation. Dev Psychol. 37:475–490 [PubMed: 11444484]

41. Aldao A (2016) Introduction to the special issue: emotion regulation as a transdiagnostic process.
Cognit Ther Res. 40:257–261
42. Aldao A, Gee DG, De Los Reyes A, Seager I (2016) Emotion regulation as a transdiagnostic
factor in the development of internalizing and externalizing psychopathology: current and future
directions. Dev Psychopathol. 28:927–946 [PubMed: 27739387]
43. Beauchaine TP, Cicchetti D (2019) Emotion dysregulation and emerging psychopathology:
A transdiagnostic, transdisciplinary perspective. Dev. Psychopathol 31:799–804 [PubMed:
31290735]
44. Cole PM, Dennis TA, Smith‐Simon KE, Cohen LH (2009) Preschoolers’ emotion regulation
strategy understanding: Relations with emotion socialization and child self‐regulation. Soc Dev
18:324–352
45. Frick PJ, Matlasz TM. Disruptive, impulse-control, and conduct disorders. In: Martel M, ed.
Developmental pathways to disruptive, impulse-control and conduct disorders. Massachusetts:
Author Manuscript

Academic Press; 2018, p. 3–20

46. Mitchison GM, Liber JM, Hannesdottir DK, Njardvik U (2020) Emotion dysregulation, ODD and
conduct problems in a sample of five and six-year-old children. Child Psychiat Hum D 51:71–79
47. Morris AS, Criss MM, Silk JS, Houltberg BJ (2017) The impact of parenting on emotion
regulation during childhood and adolescence. Child Dev. Perspect 11:233–238
48. Thompson SF, Zalewski M, Kiff CJ, Moran L, Cortes R, Lengua LJ (2020) An empirical test of
the model of socialization of emotion: Maternal and child contributors to preschoolers’ emotion
knowledge and adjustment. Dev. Psychol 56:418–430 [PubMed: 32077714]
49. Cummings EM, Davies PT (1996) Emotional security as a regulatory process in normal
development and the development of psychopathology. Dev Psychopathol 8:123–139
50. Bridgett DJ, Burt NM, Edwards ES, Deater-Deckard K (2015) Intergenerational transmission of
self-regulation: A multidisciplinary review and integrative conceptual framework. Psychol. Bull
141:602–654. [PubMed: 25938878]
51. Colman RA, Hardy SA, Albert M, Raffaelli M, Crockett L (2006) Early predictors of self-
Author Manuscript

regulation in middle childhood. Infant Child Dev 15:421–437

52. Von Suchodoletz A, Trommsdorff G, Heikamp T (2011) Linking maternal warmth and
responsiveness to children’s self‐regulation. Soc. Dev 2:486–503.
53. Parke RD, Cassidy J, Burks VM, Carson JL, Boyum L. Familial contribution to peer competence
among young children: The role of interactive and affective processes. In Parke RD, Ladd GW,
eds. Family-peer relationships: Modes of linkage. New Jersey: Lawrence Erlbaum Associates;
1992. p. 107–134
54. Morris AS, Silk JS, Steinberg L, Myers SS, Robinson LR (2007) The role of the family context in
the development of emotion regulation. Soc Dev 16:361–388 [PubMed: 19756175]
55. Duncombe ME, Havighurst SS, Holland KA, Frankling EJ (2012) The contribution of parenting
practices and parent emotion factors in children at risk for disruptive behavior disorders. Child
Psychiatry Hum. Dev 43:715–733 [PubMed: 22392414]
56. Conduct Problems Prevention Research Group. The Fast Track program for children at risk:
Preventing antisocial behavior. New York, NY: Guilford Press; 2019
57. Werthamer-Larsson L, Kellam SG, Wheeler L (1991) Effect of first grade classroom environment
Author Manuscript

on shy behavior, aggressive behavior, and concentration problems. Am J Community Psychol

19:585–602 [PubMed: 1755437]
58. Achenbach TM. Manual for the Child Behavior Checklist/4–18 and 1991 Profile. Vermont:
University of Vermont Department of Psychiatry; 1991
59. Hollingshead AA. Four-factor index of social status. New Haven, CT: Yale University; 1979
(Unpublished manuscript)
60. Jones D, Dodge KA, Foster EM, Nix R, Conduct Problems Prevention Research Group (2002)
Early identification of children at risk for costly mental health service use. Prev Science 3: 247–
256

Child Psychiatry Hum Dev. Author manuscript; available in PMC 2022 December 01.
 Yu et al. Page 20

61. McMahon RJ, Estes A. Fast Track Parent-Child Interaction Task: Observational data collection
manuals. Seattle, WA: University of Washington; 1994. (Unpublished manuscript)
Author Manuscript

62. Crnic KA, Greenberg MT (1990) Minor parenting stresses with young children. Child Dev
61:1628–1637 [PubMed: 2245752]
63. Dodge KA, Bates JE, Pettit GS (1990) Mechanisms in the cycle of violence. Science 250:1678–
1683 [PubMed: 2270481]
64. Conduct Problems Prevention Research Group. Psychometric properties of the Social Competence
Scale – Teacher and parent ratings (Fast Track Project Technical Report). Pennsylania:
Pennsylvania State University; 1995
65. Gouley KK, Brotman LM, Huang KY, Shrout PE (2008) Construct validation of the social
competence scale in preschool‐age children. Soc Dev. 17:380–398
66. Bierman KL, Kalvin CB, Heinrichs BS (2015) Early childhood precursors and adolescent sequelae
of grade school peer rejection and victimization. J Clin Child Adolesc. 44:367–379
67. Sasser TR, Kalvin CB, Bierman KL (2016) Developmental trajectories of clinically significant
attention-deficit/hyperactivity disorder (ADHD) symptoms from grade 3 through 12 in a high-risk
sample: Predictors and outcomes. J Abnorm Psychol. 125:207–219 [PubMed: 26854506]
Author Manuscript

68. Shaffer D, Fisher P. NIMH - Diagnostic Interview Schedule for Children: Parent Informant
(Interview about Child). New York: New York State Psychiatric Institute; 1991
69. Herzhoff K, Tackett JL (2016) Subfactors of oppositional defiant disorder: converging evidence
from structural and latent class analyses. J Child Psychol Psychiatry. 57:18–29 [PubMed:
25907358]
70. Ollendick TH, Booker JA, Ryan S, Greene RW (2018) Testing multiple conceptualizations of
oppositional defiant disorder in youth. J. Clin. Child Adolesc 47:620–633
71. IBM Corp (2019) IBM SPSS statistics for windows, version 26.0. IBM Corp, Armonk
72. Muthén LK, Muthén BO (1998–2013) Mplus user’s guide (7th ed). Muthén & Muthén, Los
Angeles
73. Rubin DB, Little RJ. Statistical analysis with missing data. Hoboken, NJ: Wiley; 2002
74. Hu LT, Bentler PM (1999) Cutoff criteria for fit indexes in covariance structure analysis:
Conventional criteria versus new alternatives. Struct Equ Modeling 6:1–55
75. Muthén B, Curran P (1997) General longitudinal modeling of individual differences in
Author Manuscript

experimental designs: A latent variable framework for analysis and power estimation. Psychol
Methods 2:371–402
76. MacKinnon DP, Lockwood CM, Hoffman JM, West SG, Sheets V (2002) A comparison of
methods to test mediation and other intervening variable effects. Psychol Methods 7:83–104
[PubMed: 11928892]
77. Maassen GH, & Bakker AB (2001) Suppressor variables in path models: Definitions and
interpretations. Sociol Methods Res 30:241–270
78. Morris AS, Robinson LR, Eisenberg N. Applying a multimethod perspective to the study of
developmental psychology. In: Eid M, Diener E, eds. Handbook of multimethod measurement in
psychology. Washington, DC: American Psychological Association; 2006. p. 371–384
79. Adrian M, Zeman J, Veits G (2011) Methodological implications of the affect revolution: A
35-year review of emotion regulation assessment in children. J Exp Child Psychol 110:171–197
[PubMed: 21514596]
80. Zeman J, Shipman K, Penza-Clyve S (2001) Development and initial validation of the children’s
Author Manuscript

sadness management scale. J Nonverbal Behav. 25:187–205

81. Zeman JL, Cassano M, Suveg C, Shipman K (2010) Initial validation of the children’s worry
management scale. J Child Fam Stud. 19:381–392
82. Shields A, Cicchetti D (1997) Emotion regulation among school-age children: the development and
validation of a new criterion Q-sort scale. Dev Psychol. 33:906–916 [PubMed: 9383613]
83. Eisenberg N, Spinrad TL, Fabes RA, Reiser M, Cumberland A, Shepard SA et al. (2004) The
relations of effortful control and impulsivity to children’s resiliency and adjustment. Child Dev
75:25–46 [PubMed: 15015673]

Child Psychiatry Hum Dev. Author manuscript; available in PMC 2022 December 01.
 Yu et al. Page 21

84. Hart SA, Little C, Van Bergen E (2021) Nurture might be nature: Cautionary tales and proposed
solutions. NPJ Sci Learn. 1:1–12
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript

Child Psychiatry Hum Dev. Author manuscript; available in PMC 2022 December 01.
 Yu et al. Page 22

Table 1

ODD module items in the Diagnostic Interview Schedule for Children (C-DISC) – Parent Report
Author Manuscript

Angry/Irritable Mood
1. Often loses temper
2. Often grouchy or easily annoyed
3. Often seems angry or resentful

Argumentative/Defiant Behavior
4. Often talks back or argues with adults
5. Often bothers others on purpose
6. Often blames others for their own mistakes
7. Often does things they were told not to
8. Often refuses to do as told
Author Manuscript

Spitefulness/Vindictiveness
9. Often tries to get others in trouble
10. Often gets even with other people
11. Often mean and nasty to others

Note. In order to ensure that each of the 8 symptoms comprising the diagnostic criteria of ODD in the DSM-5 were equally sampled by the
measure, items 7 and 8 were combined and items 9, 10, and 11 were combined.
Author Manuscript
Author Manuscript

Child Psychiatry Hum Dev. Author manuscript; available in PMC 2022 December 01.
 Author Manuscript Author Manuscript Author Manuscript Author Manuscript

Table 2

Descriptive statistics and correlations between study variables

Yu et al.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Covariates
1. K Initial Risk Score
2. K Sex .29**
3. K Socioeconomic Status −.13** .03

4. K Urban Black .01 .02 −.20**

5. K Urban White −.023 −.01 −.12** −.56**
6. K Rural White .00 −.21 .10** −.58** −.35**
Variables
7. K Warmth −.02 −.02 .28** −.23** .20** .06

8. G1 Warmth −.12** .00 .35** −.16** .15** .04 .42**

9. G2 Warmth −.18** .04 .28** −.13** .29** −.13** .35** .49**
10. K Harsh Discipline .14** .00 −.16** .05 −.16** −.03 −.15** −.21** −.18**
11. G1 Harsh Discipline .10* .06 −.15** .13** −.12** −.02 −.21** −.15** −.12** .42**
12. G2 Harsh Discipline .08* −.00 −.08* .17** −.08* −.11** −.16** −.15** −.13** .30** .35**
13. G2 Emotion Regulation −.43** −.24** .14** −.28** .13** .19** .13** .16** .14** −.13** −.12** −.11**
14. G3 ODD Total .32** .15** −.05 −.10* .08 .04 −.01 .00 −.06 .09* .07 .02 −.23**
15. G3 ODD Anger .27** .14** −.04 −.11** .08 .05 .02 .03 −.05 .12** .08* .07 −.18** .90**
16. G3 ODD Defiance .33** .15** −.04 −.09* .09* .02 −.02 −.01 −.06 .07 .06 −.01 −.24** .93** .69**
17. G3 ODD Spitefulness .20** .08* −.07 −.03 −.01 .04 −.02 −.04 −.06 .05 .04 .01 −.13** .76** .67** .60**

Child Psychiatry Hum Dev. Author manuscript; available in PMC 2022 December 01.
Descriptives
M 1.01 .58 25.65 .48 .26 .27 3.52 3.66 3.65 .21 .19 .12 2.00 .10 .09 .10 .07
SD 1.64 .49 12.90 .50 .44 .44 .79 .82 .84 .23 .22 .15 .76 .22 .24 .24 .26
Range −3–5 0–1 5–66 0–1 0–1 0–1 1–5 1–5 1–5 0–2 0–1 0–1 0–4 0–1 0–1 0–1 0–1

Note. K = kindergarten, G = grade, ODD = oppositional defiant disorder.

*
p < .05;
Page 23
 Yu et al. Page 24

p < .001.
p < .01;
***
**
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript

Child Psychiatry Hum Dev. Author manuscript; available in PMC 2022 December 01.
 Yu et al. Page 25

Table 3

Estimates between study variables

Author Manuscript

Variable B(SE) β p

Parental Warmth
Initial Risk −.05(.01) −.15 <.001
Sex .03(.04) .06 .493
SES .02(.00) .44 <.001
Urban Black −.04(.05) −.10 .353
Urban White .18(.05) .40 <.001
Harsh Discipline
Initial Risk .01(.01) .16 .004
Sex .01(.01) .08 .443
SES −.00(.00) −.27 <.001
Author Manuscript

Urban Black .01(.02) .05 .653

Urban White −.03(.01) −.26 .021

Emotion Regulation
Initial Risk −22(.02) −.43 <.001
Sex −.19(.05) −.19 <.001
SES .00(.00) .08 .051
Urban Black −.35(.06) −.35 <.001
Urban White −.08(.06) −.08 .205
Parental Warmth .09(.09) .05 .341
Harsh Discipline −.22(.27) −.04 .402

ODD Total
Author Manuscript

Initial Risk .02(.01) .19 .001

Sex .01(.01) .01 .539
SES .00(.00) −.01 .850
Urban Black −.03(.01) −.03 .025
Urban White −.00(.01) −.00 .803
Parental Warmth .02(.02) .07 .302
Harsh Discipline .09(.07) .09 .173
Emotion Regulation −.03(.01) −.18 .001

Anger
Initial Risk .02(.01) .16 .006
Sex .01(.01) .01 .548
SES .00(.00) .01 .856
Author Manuscript

Urban Black .02(.01) .16 .003

Urban White −.01(.02) −.01 .582
Parental Warmth .03(.02) .08 .261
Harsh Discipline .12(.08) .10 .141

Child Psychiatry Hum Dev. Author manuscript; available in PMC 2022 December 01.
 Yu et al. Page 26

Variable B(SE) β p

Emotion Regulation −.03(.01) −.16 .005

Author Manuscript

Defiance
Initial Risk .02(.01) .18 .006
Sex .00(.01) .00 .823
SES .00(.00) −.02 .622
Urban Black −.01(.01) −.01 .376
Urban White .01(.01) .01 .431
Parental Warmth .02(.02) .07 .289
Harsh Discipline .10(.08) .09 .200
Emotion Regulation −.04(.01) −.20 .001

Spitefulness
Initial Risk .02(.01) .14 .009
Author Manuscript

Sex .01(.01) .02 .251

SES .00(.00) .00 .994
Urban Black −.04(.02) −.04 .063
Urban White −.04(.02) −.04 .040
Parental Warmth −.01(.03) −.02 .745
Harsh Discipline −.01(.09) −.01 .895
Emotion Regulation −.01(.01) −.04 .456

Note. Sex: 1 = male, 0 = female; Urban Black: 1 = urban Black, 0 = rural White; urban White: 1 = urban White, 0 = rural White. Parental warmth
and harsh discipline represent latent factors. ODD total and ODD dimensions were analyzed through two separate models; estimates for parental
warmth, harsh discipline, and emotion regulation were equivalent across both models and so were only included once.
Author Manuscript
Author Manuscript

Child Psychiatry Hum Dev. Author manuscript; available in PMC 2022 December 01.
 Yu et al. Page 27

Table 4

Estimates of the Indirect Effects

Author Manuscript

Variable B SE 95% CI

ODD Total
Parental Warmth −.00 .00 −.009, .002
Harsh Discipline .01 .01 −.007, .022

Anger
Parental Warmth −.00 .00 −.009, .002
Harsh Discipline .01 .01 −.008, .024

Defiance
Parental Warmth −.00 .00 −.010, .002
Harsh Discipline .01 .01 −.008, .025
Author Manuscript

Spitefulness
Parental Warmth −.00 −.00 −.004, .002
Harsh Discipline .00 .01 −.005, .011
Author Manuscript
Author Manuscript

Child Psychiatry Hum Dev. Author manuscript; available in PMC 2022 December 01.